Error 604 in kwqmnpartition(), aborting txn after disaster recovery

Question

on a test system i test disaster recovery
szenatio:
due to an unrecoverable hardware defekt the productive server is down
rman backup sets are available
archive logs are availabe
backup hardware ready
productive system windows 2008R1 x64
backup system Win7 x32
i did the following steps to recover on the backup system:
install the same (x32) version of the rdbms than on the productive system, same directory structure
used oradim to create an instance
manually created the necessary directories "oradata" "flash_recovery" etc.
followed the steps described in the backup an recovery users guide "Restoring a Database on a New Host"
c:>RMAN nocatalog;
RMAN>connect target /
RMAN> set DBID 1249533929;
RMAN> startup nomount;
RMAN> run
RMAN> {
RMAN> ALLOCATE CHANNEL c1 DEVICE TYPE disk;
RMAN> SET CONTROLFILE AUTOBACKUP FORMAT FOR DEVICE TYPE DISK TO 'C:\Daten\Integra\rman\%F';
RMAN> restore spfile
RMAN> TO PFILE 'C:\app\administrator\product\11.1.0\db_1\database\initorcl.ora'
RMAN> from autobackup;
RMAN> shutdown abort;
RMAN> }
-- edit pfile as needed--
RMAN> startup nomount pfile='C:\app\administrator\product\11.1.0\db_1\database\INITorcl.ORA';
RMAN> run
RMAN> {
RMAN> ALLOCATE CHANNEL c1 DEVICE TYPE disk;
RMAN> SET CONTROLFILE AUTOBACKUP FORMAT FOR DEVICE TYPE DISK TO 'C:\Daten\Integra\rman\%F';
RMAN> RESTORE CONTROLFILE FROM AUTOBACKUP;
RMAN> ALTER DATABASE MOUNT;
RMAN> }
RMAN> crosscheck backup;
RMAN> restore database;
RMAN> run
2> {
3> set until sequence 672;
4> recover database;
5> }
untill this point no errors. next step would be to open the database. this is the point where trouble starts:
Datenbank ge÷ffnet
RMAN-06900: WARNUNG: Zeile V$RMAN_STATUS oder V$RMAN_OUTPUT kann nicht generiert werden
RMAN-06901: WARNUNG: Aktualisierung der Zeilen V$RMAN_STATUS und V$RMAN_OUTPUT wird deaktiviert
ORACLE-Fehler aus Zieldatenbank:
ORA-00604: Fehler auf rekursiver SQL-Ebene 1
ORA-06553: PLS-801: Interner Fehler [56319]
ORA-06553: PLS-801: Interner Fehler [56319]
looking into the alert log i find lots of errors like this:
Errors in file c:\app\administrator\diag\rdbms\orcl\orcl\trace\orcl_ora_4080.trc:
ORA-00313: Member von Log-Gruppe 1 (Thread 1) konnten nicht geöffnet werden
ORA-00312: Online-Log 1, Thread 1: 'C:\APP\ADMINISTRATOR\ORADATA\ORCL\REDO01.LOG'
ORA-27041: Datei kann nicht geöffnet werden
OSD-04002: Datei kann nicht geöffnet werden
O/S-Error: (OS 2) Das System kann die angegebene Datei nicht finden.
further down i found:
Completed: alter database open resetlogs
Wed Nov 03 15:14:02 2010
Errors in file c:\app\administrator\diag\rdbms\orcl\orcl\trace\orcl_j005_2304.trc:
ORA-12012: error on auto execute of job 55875
ORA-06553: PLS-801: internal error [56319]
Wed Nov 03 15:14:02 2010
Errors in file c:\app\administrator\diag\rdbms\orcl\orcl\trace\orcl_j002_3020.trc:
ORA-12012: error on auto execute of job 55874
ORA-06553: PLS-801: internal error [56319]
Wed Nov 03 15:14:02 2010
the instance is up, i can connect with pl/sql
the complete alert log can be posted if necessary.
is there something wrong with my rman backup sets or in the aproach for a recovery in general?
any help is welcome

Answer 1

Your source database was 64bit. You are recovering to a 32bit platform.
The wordsize has to be changed.
After the RECOVER, you must STARTUP UPGRADE and run $ORACLE_HOME/rdbms/admin/utlirp.sql
See Oracle Support Note 62290.1
Hemant K Chitale
http://hemantoracledba.blogspot.com